INDEX

Explanations

phrases and words that indicate comparisons and contrasts in opinions or analyses

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nev

-0.15

ente

-0.14

Ø®Ø§Øµ

-0.14

ÑĢÑı

-0.14

rais

-0.14

 ÑĢÐ°Ð·Ñĸ

-0.14

.study

-0.14

apur

-0.13

ãģ£ãģı

-0.13

leyen

-0.13

POSITIVE LOGITS

 meanwhile

0.72

 Meanwhile

0.65

Meanwhile

0.63

 meantime

0.47

 convers

0.46

 Conversely

0.46

 whereas

0.44

 naopak

0.36

 Whereas

0.36

 contrast

0.35

Activations Density 0.269%

phrases and words that indicate comparisons and contrasts in opinions or analyses

No Comments

No Known Activations